A combined model using this muscle index and swallowing test results improved prediction performance to an acceptable level.
Medicine · 8 authors, 2 centres
This summary was generated by AI from a single paper. It has not been reviewed by a clinician and is not clinical advice. Verify against the source before acting on it.
A combined model using this muscle index and swallowing test results improved prediction performance to an acceptable level.
This longitudinal, observational study included 146 patients with aspiration pneumonia from a single center. The primary outcome was successful weaning from nasogastric tube feeding during hospitalization, achieved in 46.6% of patients. A board-certified radiologist and a validated convolutional neural network model measured muscle and fat parameters from chest CT. Neither skeletal muscle parameters alone nor VFSS results alone were optimal for prediction, but a model combining the skeletal muscle index at the T4 level and VFSS results achieved an acceptable predictive performance (AUC ≥ 0.7). Limitations include the retrospective design, single-center data, relatively small sample size, and lack of muscle strength measurement. The findings suggest that thoracic muscle mass assessment may be a useful supplementary tool for predicting weaning success in this population.